import numpy as np
import pandas as pd
from itertools import cycle
from scipy import interp

# sklearn
from sklearn import metrics
from sklearn.model_selection import train_test_split, StratifiedShuffleSplit
from sklearn import datasets
from sklearn.preprocessing import StandardScaler
from sklearn import preprocessing

# TensorFlow
import tensorflow as tf

# Visualisation libraries

## Text
from colorama import Fore, Back, Style
from IPython.display import Image, display, Markdown, Latex, clear_output

## progressbar
import progressbar

## plotly
from plotly.offline import init_notebook_mode, iplot 
import plotly.graph_objs as go
import plotly.offline as py
from plotly.subplots import make_subplots
import plotly.express as px

## seaborn
import seaborn as sns

## matplotlib
import matplotlib.pyplot as plt
from matplotlib.patches import Ellipse, Polygon
from matplotlib.font_manager import FontProperties
import matplotlib.colors as mcolors
from matplotlib import cm
plt.style.use('seaborn-whitegrid')
plt.rcParams['axes.labelsize'] = 14
plt.rcParams['xtick.labelsize'] = 12
plt.rcParams['ytick.labelsize'] = 12
plt.rcParams['text.color'] = 'k'
%matplotlib inline

import warnings
warnings.filterwarnings("ignore")

Home / Deep Learning

Tensorflow Classifiers

• Tensorflow Linear Classifier
• Tensorflow DNN Classifier
• Tensorflow Boosted Trees Classifier
• Tensorflow Boosted Trees Classifier with Feature Importance Analysis

In this article, we demonstrate solving a classification problem in TensorFlow using Estimators using the UCI ML Wine recognition dataset. This dataset also can be accessed via the scikit-learn datasets.

Dataset Information:

These data are the results of a chemical analysis of wines grown in the same region in Italy but derived from three different cultivars. The analysis determined the quantities of 13 constituents found in each of the three types of wines.

data = datasets.load_wine()
Data = pd.DataFrame(data['data'],
                    columns = [x.title().replace('_',' ').replace('Of','of').replace('Od','OD') for x in data['feature_names']])
Labels_dict = dict(zip(list(np.sort(np.unique(data['target'].tolist()))),
                       list([x.title() for x in data['target_names']])))
Target = 'Class'
Data[Target] = data['target']
display(Data)
print(data['DESCR'])

	Alcohol	Malic Acid	Ash	Alcalinity of Ash	Magnesium	Total Phenols	Flavanoids	Nonflavanoid Phenols	Proanthocyanins	Color Intensity	Hue	OD280/OD315 of Diluted Wines	Proline	Class
0	14.23	1.71	2.43	15.6	127.0	2.80	3.06	0.28	2.29	5.64	1.04	3.92	1065.0	0
1	13.20	1.78	2.14	11.2	100.0	2.65	2.76	0.26	1.28	4.38	1.05	3.40	1050.0	0
2	13.16	2.36	2.67	18.6	101.0	2.80	3.24	0.30	2.81	5.68	1.03	3.17	1185.0	0
3	14.37	1.95	2.50	16.8	113.0	3.85	3.49	0.24	2.18	7.80	0.86	3.45	1480.0	0
4	13.24	2.59	2.87	21.0	118.0	2.80	2.69	0.39	1.82	4.32	1.04	2.93	735.0	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
173	13.71	5.65	2.45	20.5	95.0	1.68	0.61	0.52	1.06	7.70	0.64	1.74	740.0	2
174	13.40	3.91	2.48	23.0	102.0	1.80	0.75	0.43	1.41	7.30	0.70	1.56	750.0	2
175	13.27	4.28	2.26	20.0	120.0	1.59	0.69	0.43	1.35	10.20	0.59	1.56	835.0	2
176	13.17	2.59	2.37	20.0	120.0	1.65	0.68	0.53	1.46	9.30	0.60	1.62	840.0	2
177	14.13	4.10	2.74	24.5	96.0	2.05	0.76	0.56	1.35	9.20	0.61	1.60	560.0	2

178 rows × 14 columns

.. _wine_dataset:

Wine recognition dataset
------------------------

**Data Set Characteristics:**

    :Number of Instances: 178 (50 in each of three classes)
    :Number of Attributes: 13 numeric, predictive attributes and the class
    :Attribute Information:
 		- Alcohol
 		- Malic acid
 		- Ash
		- Alcalinity of ash  
 		- Magnesium
		- Total phenols
 		- Flavanoids
 		- Nonflavanoid phenols
 		- Proanthocyanins
		- Color intensity
 		- Hue
 		- OD280/OD315 of diluted wines
 		- Proline

    - class:
            - class_0
            - class_1
            - class_2
		
    :Summary Statistics:
    
    ============================= ==== ===== ======= =====
                                   Min   Max   Mean     SD
    ============================= ==== ===== ======= =====
    Alcohol:                      11.0  14.8    13.0   0.8
    Malic Acid:                   0.74  5.80    2.34  1.12
    Ash:                          1.36  3.23    2.36  0.27
    Alcalinity of Ash:            10.6  30.0    19.5   3.3
    Magnesium:                    70.0 162.0    99.7  14.3
    Total Phenols:                0.98  3.88    2.29  0.63
    Flavanoids:                   0.34  5.08    2.03  1.00
    Nonflavanoid Phenols:         0.13  0.66    0.36  0.12
    Proanthocyanins:              0.41  3.58    1.59  0.57
    Colour Intensity:              1.3  13.0     5.1   2.3
    Hue:                          0.48  1.71    0.96  0.23
    OD280/OD315 of diluted wines: 1.27  4.00    2.61  0.71
    Proline:                       278  1680     746   315
    ============================= ==== ===== ======= =====

    :Missing Attribute Values: None
    :Class Distribution: class_0 (59), class_1 (71), class_2 (48)
    :Creator: R.A. Fisher
    :Donor: Michael Marshall (MARSHALL%PLU@io.arc.nasa.gov)
    :Date: July, 1988

This is a copy of UCI ML Wine recognition datasets.
https://archive.ics.uci.edu/ml/machine-learning-databases/wine/wine.data

The data is the results of a chemical analysis of wines grown in the same
region in Italy by three different cultivators. There are thirteen different
measurements taken for different constituents found in the three types of
wine.

Original Owners: 

Forina, M. et al, PARVUS - 
An Extendible Package for Data Exploration, Classification and Correlation. 
Institute of Pharmaceutical and Food Analysis and Technologies,
Via Brigata Salerno, 16147 Genoa, Italy.

Citation:

Lichman, M. (2013). UCI Machine Learning Repository
[https://archive.ics.uci.edu/ml]. Irvine, CA: University of California,
School of Information and Computer Science. 

.. topic:: References

  (1) S. Aeberhard, D. Coomans and O. de Vel, 
  Comparison of Classifiers in High Dimensional Settings, 
  Tech. Rep. no. 92-02, (1992), Dept. of Computer Science and Dept. of  
  Mathematics and Statistics, James Cook University of North Queensland. 
  (Also submitted to Technometrics). 

  The data was used with many others for comparing various 
  classifiers. The classes are separable, though only RDA 
  has achieved 100% correct classification. 
  (RDA : 100%, QDA 99.4%, LDA 98.9%, 1NN 96.1% (z-transformed data)) 
  (All results using the leave-one-out technique) 

  (2) S. Aeberhard, D. Coomans and O. de Vel, 
  "THE CLASSIFICATION PERFORMANCE OF RDA" 
  Tech. Rep. no. 92-01, (1992), Dept. of Computer Science and Dept. of 
  Mathematics and Statistics, James Cook University of North Queensland. 
  (Also submitted to Journal of Chemometrics).

Features with high variance

Moreover, high variance for some features can hurt our modeling process. For this reason, we would like to standardize features by removing the mean and scaling to unit variance.

def Feature_Normalize(X, PD):
    def List_Break(mylist, n = PD['word_break']):
        Out = []
        for x in mylist:
            y = x.split()
            if len(y)> n:
                z = ' '.join(y[:n])
                sep = np.arange(0, len(y), n)[1:]
                for n in sep:
                    z = z + '\n'+ ' '.join(y[n:])
            else:
                z = ' '.join(y)
            Out.append(z)
        return Out

    X = Data.drop(columns = [Target])
    y = Data[Target]

    scaler = preprocessing.StandardScaler()
    X_std = scaler.fit_transform(X)
    X_std = pd.DataFrame(data = X_std, columns = X.columns)

    fig, ax = plt.subplots(2, 1, figsize = PD['figsize'])
    ax = ax.ravel()
    CP = [sns.color_palette("OrRd", 20), sns.color_palette("Greens", X.shape[1])]
    Names = ['Variance of the Features', 'Variance of the Features (Standardized)']
    Sets = [X, X_std]
    kws = dict(label='Feature\nVariance', aspect=10, shrink= .3)
    for i in range(len(ax)):
        Temp = Sets[i].var().sort_values(ascending = False).to_frame(name= 'Variance').round(2).T
        _ = sns.heatmap(Temp, ax=ax[i], annot=True, square=True,  cmap = CP[i],
                        linewidths = 0.8, vmin=0, vmax=Temp.max(axis =1)[0],  annot_kws={"size": PD['annot_text_size']},
                        cbar_kws=kws)
        if not PD['word_break'] == None:
            mylist = List_Break(Temp.T.index.tolist())
            _ = ax[i].xaxis.set_ticklabels(mylist)
        _ = ax[i].set_yticklabels('')
        _ = ax[i].set_title(Names[i], weight='bold', fontsize = 14)
        _ = ax[i].set_aspect(1)
        del Temp
    plt.subplots_adjust(hspace=PD['hspace'])
    Out = pd.DataFrame(data = X_std, columns = X.columns.tolist())
    return Out

X = Data.drop(columns = [Target])
y = Data[Target]

PD = dict(figsize = (16, 8), hspace = 0.2, annot_text_size = 12, word_break = 2)
X = Feature_Normalize(X, PD)

Train and Test sets

def DatasetTargetDist(Inp, Target, Labels_dict, PD):
    # Table
    Table = Inp[Target].value_counts().to_frame('Count').reset_index(drop = False).rename(columns = {'index':Target})
    Table[Target] = Table[Target].replace(Labels_dict)
    Table['Percentage'] = np.round(100*(Table['Count']/Table['Count'].sum()),2)
    fig = make_subplots(rows=1, cols=2, horizontal_spacing = 0.02, column_widths=PD['column_widths'],
                        specs=[[{"type": "table"},{"type": "pie"}]])

    # Right
    fig.add_trace(go.Pie(labels=Table[Target].values, values=Table['Count'].values,
                         pull=PD['pull'], textfont=dict(size= PD['textfont']),
                         marker=dict(colors = PD['PieColors'], line=dict(color='black', width=1))), row=1, col=2)
    fig.update_traces(hole=PD['hole'])
    fig.update_layout(height = PD['height'], legend=dict(orientation="v"), legend_title_text= PD['legend_title'])
    
    # Left
    T = Table.copy()
    T['Percentage'] = T['Percentage'].map(lambda x: '%%%.2f' % x)
    Temp = []
    for i in T.columns:
        Temp.append(T.loc[:,i].values)

    fig.add_trace(go.Table(header=dict(values = list(Table.columns), line_color='darkslategray',
                                       fill_color= PD['TableColors'][0], align=['center','center'],
                                       font=dict(color='white', size=12), height=25), columnwidth = PD['tablecolumnwidth'],
                           cells=dict(values=Temp, line_color='darkslategray',
                                      fill=dict(color= [PD['TableColors'][1], PD['TableColors'][1]]),
                                  align=['center', 'center'], font_size=12, height=20)), 1, 1)

    fig.update_layout(title={'text': '<b>' + Target + '<b>', 'x':PD['title_x'],
                                     'y':PD['title_y'], 'xanchor': 'center', 'yanchor': 'top'})
    fig.show()

Pull = [0.02 for x in range((len(Labels_dict)-1))]
Pull.append(.05)
PD = dict(PieColors = px.colors.sequential.BuPu[1:-1:3],
          TableColors = ['Navy','White'], hole = .4,
          column_widths=[0.6, 0.4],textfont = 14, height = 400, tablecolumnwidth = [0.20, 0.12, 0.15],
          pull = Pull, legend_title = Target, title_x = 0.5, title_y = 0.8)
del Pull
DatasetTargetDist(Data, Target, Labels_dict, PD)

StratifiedKFold is a variation of k-fold which returns stratified folds: each set contains approximately the same percentage of samples of each target class as the complete set.

Test_Size = 0.3
sss = StratifiedShuffleSplit(n_splits=1, test_size=Test_Size, random_state=42)
_ = sss.get_n_splits(X, y)
# For Tensorflow
X.columns = [x.replace(' ','_') for x in X.columns]

for train_index, test_index in sss.split(X, y):
    # X
    if isinstance(X, pd.DataFrame):
        X_train, X_test = X.loc[train_index], X.loc[test_index]
    else:
        X_train, X_test = X[train_index], X[test_index]
    # y    
    if isinstance(y, pd.Series):
        y_train, y_test = y[train_index], y[test_index]
    else:
        y_train, y_test = y[train_index], y[test_index]
del sss

def Train_Test_Dist(X_train, y_train, X_test, y_test, PD, Labels_dict = Labels_dict):
    def ToSeries(x):
        if not isinstance(x, pd.Series):
            Out = pd.Series(x)
        else:
            Out = x.copy()
        return Out
    
    fig = make_subplots(rows=1, cols=3, horizontal_spacing = 0.02, column_widths= PD['column_widths'],
                        specs=[[{"type": "table"},{'type':'domain'}, {'type':'domain'}]])
    # Right
    C = 2
    for y in [ToSeries(y_train).replace(Labels_dict), ToSeries(y_test).replace(Labels_dict)]:
        fig.add_trace(go.Pie(labels= list(Labels_dict.values()),
                             values= y.value_counts().values, pull=PD['pull'],
                             textfont=dict(size=PD['textfont']),
                             marker=dict(colors = PD['PieColors'],
                             line=dict(color='black', width=1))), row=1, col=C)
        fig.update_traces(hole=.5)
        fig.update_layout(legend=dict(orientation="v"), legend_title_text= PD['legend_title'])
        C+=1
    
    # Left
    
    # Table
    Table = pd.DataFrame(data={'Set':['X_train','X_test','y_train','y_test'],
                       'Shape':[X_train.shape, X_test.shape, y_train.shape, y_test.shape]}).astype(str)
    T = Table.copy()
    Temp = []
    for i in T.columns:
        Temp.append(T.loc[:,i].values)
        
    TableColors = PD['TableColors']
    fig.add_trace(go.Table(header=dict(values = list(Table.columns), line_color='darkslategray',
                                       fill_color= TableColors[0], align=['center','center'],
                                       font=dict(color='white', size=12), height=25), columnwidth = PD['tablecolumnwidth'],
                           cells=dict(values=Temp, line_color='darkslategray',
                                      fill=dict(color= [TableColors[1], TableColors[1]]),
                                  align=['center', 'center'], font_size=12, height=20)), 1, 1)
    fig.update_layout(title={'text': '<b>' + 'Dataset Distribution' + '<b>', 'x':PD['title_x'],
                                     'y':PD['title_y'], 'xanchor': 'center', 'yanchor': 'top'})
    if not PD['height'] == None:
        fig.update_layout(height = PD['height'])
    fig.show()
    
PD.update(dict(column_widths=[0.3, 0.3, 0.3], tablecolumnwidth = [0.2, 0.4], height = 350, legend_title = Target))

Train_Test_Dist(X_train, y_train, X_test, y_test, PD)

Modeling: Tensorflow DNNClassifier

Here, we use the Tensorflow Linear classifier model.tf.estimator.DNNClassifier.

Input Function

The input function specifies how data is converted to a tf.data.Dataset that feeds the input pipeline in a streaming fashion. Moreover, an input function is a function that returns a tf.data.Dataset object which outputs the following two-element tuple:

• features - A Python dictionary in which:
  • Each key is the name of a feature.
  • Each value is an array containing all of that feature's values.
• label - An array containing the values of the label for every example.

def input_fn(features, labels, training=True, batch_size=256):
    """An input function for training or evaluating"""
    # Convert the inputs to a Dataset.
    dataset = tf.data.Dataset.from_tensor_slices((dict(features), labels))

    # Shuffle and repeat if you are in training mode.
    if training:
        dataset = dataset.shuffle(1000).repeat()
    
    return dataset.batch(batch_size)

Moreover, an estimator model consists of two main parts, feature columns, and a numeric vector. Feature columns provide explanations for the input numeric vector. The following function separates categorical and numerical columns (features)and returns a descriptive list of feature columns.

def Feat_Columns(Inp):
    Temp = Inp.dtypes.reset_index(drop = False)
    Temp.columns = ['Features', 'Data Type']
    Temp['Data Type'] = Temp['Data Type'].astype(str)
    # Numeric_Columns
    Numeric_Columns = Temp.loc[Temp['Data Type'].isin(['int64', 'int32', 'float64', 'float32']),'Features'].tolist()
    # Categorical_Columns
    Categorical_Columns = Temp.loc[Temp['Data Type'] == 'object','Features'].tolist()
        
    # Feature Columns
    feature_columns = []
    if len(Categorical_Columns)>0:
        for feature_name in Categorical_Columns:
          vocabulary = Inp[feature_name].unique()
          feature_columns.append(tf.feature_column.categorical_column_with_vocabulary_list(feature_name, vocabulary))
    if len(Numeric_Columns)>0:
        for feature_name in Numeric_Columns:
          feature_columns.append(tf.feature_column.numeric_column(feature_name))
    return feature_columns

my_feature_columns = Feat_Columns(X)

Estimator using the Default Optimizer

tf.keras.backend.clear_session()
IT = int(5e3)
classifier = tf.estimator.DNNClassifier(feature_columns=my_feature_columns,
                                        # Hidden layers.
                                        hidden_units= [100, 12, 3],
                                        # The model must choose between 3 classes.
                                        n_classes=len(Labels_dict))
classifier.train(input_fn=lambda: input_fn(X_train, y_train, training=True), max_steps = IT)
result = classifier.evaluate(input_fn=lambda: input_fn(X_test, y_test, training=False))
clear_output()
display(pd.DataFrame(result, index = ['']).round(4))

	accuracy	average_loss	loss	global_step
	0.6296	0.7495	0.7495	5000

ROC Curves

def ROC_Curve(y_test, probs, n_classes, FS = 7, ax = False, pad = 0.01):
    # converting y_test to categorical
    y_test_cat = tf.keras.utils.to_categorical(y_test, num_classes=n_classes, dtype='float32')
    # Compute ROC curve and ROC area for each class
    fpr = dict()
    tpr = dict()
    roc_auc = dict()
    for i in range(n_classes):
        fpr[i], tpr[i], _ = metrics.roc_curve(y_test_cat[:, i], probs[:, i])
        roc_auc[i] = metrics.auc(fpr[i], tpr[i])

    # Compute micro-average ROC curve and ROC area
    fpr["micro"], tpr["micro"], _ = metrics.roc_curve(y_test_cat.ravel(), probs.ravel())
    roc_auc["micro"] = metrics.auc(fpr["micro"], tpr["micro"])

    # First aggregate all false positive rates
    all_fpr = np.unique(np.concatenate([fpr[i] for i in range(n_classes)]))

    # Then interpolate all ROC curves at this points
    mean_tpr = np.zeros_like(all_fpr)
    for i in range(n_classes):
        mean_tpr += interp(all_fpr, fpr[i], tpr[i])

    # Finally average it and compute AUC
    mean_tpr /= n_classes

    fpr["macro"] = all_fpr
    tpr["macro"] = mean_tpr
    roc_auc["macro"] = metrics.auc(fpr["macro"], tpr["macro"])

    fig = go.Figure()
    fig.add_trace(go.Scatter(x=[0, 1], y=[0, 1], name = 'FPR = TPR', line = dict(color='Black', width=2, dash='dash')))
    fig.add_trace(go.Scatter(x=fpr["micro"], y=tpr["micro"], mode='lines', marker_color = 'deeppink',
                             name='micro-average ROC curve (area = {0:0.2f})'.format(roc_auc["micro"])))
    fig.add_trace(go.Scatter(x=fpr["macro"], y=tpr["macro"], mode='lines', marker_color = 'navy',
                             name='macro-average ROC curve (area = {0:0.2f})'.format(roc_auc["macro"])))

    colors = cycle(['Aqua', 'DarkOrange', 'CornflowerBlue'])
    for i, color in zip(range(n_classes), colors):
        _ = fig.add_trace(go.Scatter(x = fpr[i], y = tpr[i], mode='lines', marker_color= px.colors.sequential.Rainbow,
                                     name='ROC curve of class {0} (area = {1:0.2f})'.format(i, roc_auc[i])))

    # Background
    fig.update_layout(plot_bgcolor= 'white')
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=True, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range =[-pad, 1+pad],
                     title = 'False Positive Rate (FPR)')
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=True, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range =[-pad, 1+pad],
                     title = 'True Positive Rate (TPR)')
    fig.update_yaxes(scaleanchor = "x", scaleratio = 1)
    fig.update_layout(height = 600, width = 810)
    fig.update_layout(title={'text': '<b>' + 'Receiver Operating Characteristic (ROC) Curves' + '<b>', 'x': .5,
                                         'y': .9, 'xanchor': 'center', 'yanchor': 'top'})
    fig.show()

pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_test, y_test, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])    
ROC_Curve(y_test, probs, n_classes = len(Labels_dict), FS = 8)

Confusion Matrix

The confusion matrix allows for visualization of the performance of an algorithm. Note that due to the size of data, here we don't provide a Cross-validation evaluation. In general, this type of evaluation is preferred.

def Confusion_Mat(CM_Train, CM_Test, PD, n_splits = 10):
    if n_splits == None:
        Titles = ['Train Set', 'Test Set']
    else:
        Titles = ['Train Set (CV = % i)' % n_splits, 'Test Set (CV = % i)' % n_splits]
    CM = [CM_Train, CM_Test]
    Cmap = ['Greens', 'YlGn','Blues', 'PuBu']
    for i in range(2):
        fig, ax = plt.subplots(1, 2, figsize= PD['FS'])
        fig.suptitle(Titles[i], weight = 'bold', fontsize = 16)
        _ = sns.heatmap(CM[i], annot=True, annot_kws={"size": PD['annot_kws']}, cmap=Cmap[2*i], ax = ax[0],
                        linewidths = 0.2, cbar_kws={"shrink": PD['shrink']})
        _ = ax[0].set_title('Confusion Matrix');
        Temp = np.round(CM[i].astype('float') / CM[i].sum(axis=1)[:, np.newaxis], 2)
        _ = sns.heatmap(Temp,
                        annot=True, annot_kws={"size": PD['annot_kws']}, cmap=Cmap[2*i+1], ax = ax[1],
                       linewidths = 0.4, vmin=0, vmax=1, cbar_kws={"shrink": PD['shrink']})
        _ = ax[1].set_title('Normalized Confusion Matrix');

        for a in ax:
            _ = a.set_xlabel('Predicted labels')
            _ = a.set_ylabel('True labels'); 
            _ = a.xaxis.set_ticklabels(PD['Labels'])
            _ = a.yaxis.set_ticklabels(PD['Labels'])
            _ = a.set_aspect(1)

# Train
pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_train, y_train, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])
y_pred = np.argmax(probs, axis = 1).reshape(-1,1)
Reports_Train = pd.DataFrame(metrics.classification_report(y_train, y_pred, target_names=list(Labels_dict.values()),
                                                           output_dict=True)).T
CM_Train = metrics.confusion_matrix(y_train, y_pred)
# Test
pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_test, y_test, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])
y_pred = np.argmax(probs, axis = 1).reshape(-1,1)
Reports_Test = pd.DataFrame(metrics.classification_report(y_test, y_pred, target_names=list(Labels_dict.values()),
                                                          output_dict=True)).T
CM_Test = metrics.confusion_matrix(y_test, y_pred)

Reports_Train = Reports_Train.reset_index().rename(columns ={'index': 'Train Set'})
Reports_Test = Reports_Test.reset_index().rename(columns ={'index': 'Test Set'})
                                                 
display(Reports_Train.style.hide_index().set_properties(**{'background-color': 'HoneyDew', 'color': 'Black'}).\
        set_properties(subset=['Train Set'], **{'background-color': 'SeaGreen', 'color': 'White'}))
display(Reports_Test.style.hide_index().set_properties(**{'background-color': 'Azure', 'color': 'Black'}).\
        set_properties(subset=['Test Set'], **{'background-color': 'RoyalBlue', 'color': 'White'}))

PD = dict(FS = (12, 6), annot_kws = 14, shrink = .6, Labels = list(Labels_dict.values()))
Confusion_Mat(CM_Train, CM_Test, PD = PD, n_splits = None)

Train Set	precision	recall	f1-score	support
Class_0	1.000000	1.000000	1.000000	41.000000
Class_1	0.602410	1.000000	0.751880	50.000000
Class_2	0.000000	0.000000	0.000000	33.000000
accuracy	0.733871	0.733871	0.733871	0.733871
macro avg	0.534137	0.666667	0.583960	124.000000
weighted avg	0.573552	0.733871	0.633822	124.000000

Test Set	precision	recall	f1-score	support
Class_0	0.782609	1.000000	0.878049	18.000000
Class_1	0.516129	0.761905	0.615385	21.000000
Class_2	0.000000	0.000000	0.000000	15.000000
accuracy	0.629630	0.629630	0.629630	0.629630
macro avg	0.432913	0.587302	0.497811	54.000000
weighted avg	0.461586	0.629630	0.531999	54.000000

Estimator using an Optimizer with a Learning Rate Decay

In this classification, the learning rate of your optimizer changes over time.

tf.keras.backend.clear_session()
classifier = tf.estimator.DNNClassifier(feature_columns=my_feature_columns,
            # Hidden layers.
            hidden_units= [1024, 512, 256],
            # The model must choose between 3 classes.
            n_classes=len(Labels_dict),
            optimizer=lambda: tf.keras.optimizers.Adam(learning_rate=tf.compat.v1.train.exponential_decay(learning_rate=0.1,
            global_step=tf.compat.v1.train.get_global_step(), decay_steps=IT,decay_rate=0.96)))
#
classifier.train(input_fn=lambda: input_fn(X_train, y_train, training=True), max_steps= IT)
result = classifier.evaluate(input_fn=lambda: input_fn(X_test, y_test, training=False))
clear_output()
display(pd.DataFrame(result, index = ['']).round(4))

	accuracy	average_loss	loss	global_step
	0.963	105.379303	105.379303	5000

ROC Curves

pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_test, y_test, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])    
ROC_Curve(y_test, probs, n_classes = len(Labels_dict), FS = 8)

Confusion Matrix

# Train
pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_train, y_train, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])
y_pred = np.argmax(probs, axis = 1).reshape(-1,1)
Reports_Train = pd.DataFrame(metrics.classification_report(y_train, y_pred, target_names=list(Labels_dict.values()),
                                                           output_dict=True)).T
CM_Train = metrics.confusion_matrix(y_train, y_pred)
# Test
pred_dicts = list(classifier.predict(input_fn=lambda: input_fn(X_test, y_test, training=False)))
clear_output()
probs = np.array([pred['probabilities'] for pred in pred_dicts])
y_pred = np.argmax(probs, axis = 1).reshape(-1,1)
Reports_Test = pd.DataFrame(metrics.classification_report(y_test, y_pred, target_names=list(Labels_dict.values()),
                                                          output_dict=True)).T
CM_Test = metrics.confusion_matrix(y_test, y_pred)

Reports_Train = Reports_Train.reset_index().rename(columns ={'index': 'Train Set'})
Reports_Test = Reports_Test.reset_index().rename(columns ={'index': 'Test Set'})
                                                 
display(Reports_Train.style.hide_index().set_properties(**{'background-color': 'HoneyDew', 'color': 'Black'}).\
        set_properties(subset=['Train Set'], **{'background-color': 'SeaGreen', 'color': 'White'}))
display(Reports_Test.style.hide_index().set_properties(**{'background-color': 'Azure', 'color': 'Black'}).\
        set_properties(subset=['Test Set'], **{'background-color': 'RoyalBlue', 'color': 'White'}))

PD = dict(FS = (12, 6), annot_kws = 14, shrink = .6, Labels = list(Labels_dict.values()))
Confusion_Mat(CM_Train, CM_Test, PD = PD, n_splits = None)

Train Set	precision	recall	f1-score	support
Class_0	1.000000	1.000000	1.000000	41.000000
Class_1	1.000000	1.000000	1.000000	50.000000
Class_2	1.000000	1.000000	1.000000	33.000000
accuracy	1.000000	1.000000	1.000000	1.000000
macro avg	1.000000	1.000000	1.000000	124.000000
weighted avg	1.000000	1.000000	1.000000	124.000000

Test Set	precision	recall	f1-score	support
Class_0	0.900000	1.000000	0.947368	18.000000
Class_1	1.000000	0.904762	0.950000	21.000000
Class_2	1.000000	1.000000	1.000000	15.000000
accuracy	0.962963	0.962963	0.962963	0.962963
macro avg	0.966667	0.968254	0.965789	54.000000
weighted avg	0.966667	0.962963	0.963012	54.000000

---

References

1. Regression analysis Wikipedia page
2. Tensorflow tutorials
3. Online machine learning Wikipedia page
4. Learning rate Wikipedia page
5. S. Aeberhard, D. Coomans and O. de Vel, Comparison of Classifiers in High Dimensional Settings, Tech. Rep. no. 92-02, (1992), Dept. of Computer Science and Dept. of Mathematics and Statistics, James Cook University of North Queensland. (Also submitted to Technometrics).
6. S. Aeberhard, D. Coomans and O. de Vel, “THE CLASSIFICATION PERFORMANCE OF RDA” Tech. Rep. no. 92-01, (1992), Dept. of Computer Science and Dept. of Mathematics and Statistics, James Cook University of North Queensland. (Also submitted to Journal of Chemometrics).

---